💼 About This Role

You'll work with our Science team to define 'good' model behavior for Reasoning, Audio, Alignment, and Tools. You'll design evaluation pipelines and transform insights into best-in-class models.

🎯 What You'll Do

• Interact with models to identify areas for behavior improvement
• Gather internal and external feedback to scope improvements
• Design and implement evals, data guidelines, and synthetic tests
• Develop robust evaluation pipelines for model candidates

📋 Requirements

• Deep understanding of linguistics, engineering, or LLM agents
• Prior knowledge in training and optimizing model behavior
• Expertise in building robust evaluations

✨ Nice to Have

• Experience with reasoning and tool use
• Track record of innovative solutions for real-world constraints

🎁 Benefits & Perks

• 🏖️ Flexible PTO
• 💻 Remote-friendly culture
• 📚 Learning budget
• 🏥 Health insurance